Netizen marvels: AI inference burns 6.45M tokens per minute, raising concerns

huangyun_122 · x · 2026-08-08

Netizen @huangyun122, replying to @tinyfool, notes that AI inference speed reaches 6.45 million tokens per minute, meaning 6.45 million tokens are consumed each minute, a massive compute usage that feels frightening.

Original post →

More from AGI Musings

AGI Musings channel →